January 28, 2020

3169 words 15 mins read

Paper Group ANR 855

Revisiting Paraphrase Question Generator using Pairwise Discriminator. Bayesian Deconditional Kernel Mean Embeddings. Single Episode Policy Transfer in Reinforcement Learning. Making Recommendations from Web Archives for “Lost” Web Pages. Deep Reinforcement Learning for Chatbots Using Clustered Actions and Human-Likeness Rewards. Generalizable Reso …

Revisiting Paraphrase Question Generator using Pairwise Discriminator


Title	Revisiting Paraphrase Question Generator using Pairwise Discriminator
Authors	Badri N. Patro, Dev Chauhan, Vinod K. Kurmi, Vinay P. Namboodiri
Abstract	In this paper, we propose a method for obtaining sentence-level embeddings. While the problem of securing word-level embeddings is very well studied, we propose a novel method for obtaining sentence-level embeddings. This is obtained by a simple method in the context of solving the paraphrase generation task. If we use a sequential encoder-decoder model for generating paraphrase, we would like the generated paraphrase to be semantically close to the original sentence. One way to ensure this is by adding constraints for true paraphrase embeddings to be close and unrelated paraphrase candidate sentence embeddings to be far. This is ensured by using a sequential pair-wise discriminator that shares weights with the encoder that is trained with a suitable loss function. Our loss function penalizes paraphrase sentence embedding distances from being too large. This loss is used in combination with a sequential encoder-decoder network. We also validated our method by evaluating the obtained embeddings for a sentiment analysis task. The proposed method results in semantic embeddings and outperforms the state-of-the-art on the paraphrase generation and sentiment analysis task on standard datasets. These results are also shown to be statistically significant.
Tasks	Paraphrase Generation, Sentence Embedding, Sentence Embeddings, Sentiment Analysis
Published	2019-12-31
URL	https://arxiv.org/abs/1912.13149v2
PDF	https://arxiv.org/pdf/1912.13149v2.pdf
PWC	https://paperswithcode.com/paper/revisiting-paraphrase-question-generator
Repo
Framework

Bayesian Deconditional Kernel Mean Embeddings


Title	Bayesian Deconditional Kernel Mean Embeddings
Authors	Kelvin Hsu, Fabio Ramos
Abstract	Conditional kernel mean embeddings form an attractive nonparametric framework for representing conditional means of functions, describing the observation processes for many complex models. However, the recovery of the original underlying function of interest whose conditional mean was observed is a challenging inference task. We formalize deconditional kernel mean embeddings as a solution to this inverse problem, and show that it can be naturally viewed as a nonparametric Bayes’ rule. Critically, we introduce the notion of task transformed Gaussian processes and establish deconditional kernel means as their posterior predictive mean. This connection provides Bayesian interpretations and uncertainty estimates for deconditional kernel mean embeddings, explains their regularization hyperparameters, and reveals a marginal likelihood for kernel hyperparameter learning. These revelations further enable practical applications such as likelihood-free inference and learning sparse representations for big data.
Tasks	Gaussian Processes
Published	2019-06-01
URL	https://arxiv.org/abs/1906.00199v1
PDF	https://arxiv.org/pdf/1906.00199v1.pdf
PWC	https://paperswithcode.com/paper/190600199
Repo
Framework

Single Episode Policy Transfer in Reinforcement Learning


Title	Single Episode Policy Transfer in Reinforcement Learning
Authors	Jiachen Yang, Brenden Petersen, Hongyuan Zha, Daniel Faissol
Abstract	Transfer and adaptation to new unknown environmental dynamics is a key challenge for reinforcement learning (RL). An even greater challenge is performing near-optimally in a single attempt at test time, possibly without access to dense rewards, which is not addressed by current methods that require multiple experience rollouts for adaptation. To achieve single episode transfer in a family of environments with related dynamics, we propose a general algorithm that optimizes a probe and an inference model to rapidly estimate underlying latent variables of test dynamics, which are then immediately used as input to a universal control policy. This modular approach enables integration of state-of-the-art algorithms for variational inference or RL. Moreover, our approach does not require access to rewards at test time, allowing it to perform in settings where existing adaptive approaches cannot. In diverse experimental domains with a single episode test constraint, our method significantly outperforms existing adaptive approaches and shows favorable performance against baselines for robust transfer.
Tasks
Published	2019-10-17
URL	https://arxiv.org/abs/1910.07719v3
PDF	https://arxiv.org/pdf/1910.07719v3.pdf
PWC	https://paperswithcode.com/paper/single-episode-policy-transfer-in
Repo
Framework

Making Recommendations from Web Archives for “Lost” Web Pages


Title	Making Recommendations from Web Archives for “Lost” Web Pages
Authors	Lulwah M. Alkwai, Michael L. Nelson, Michele C. Weigle
Abstract	When a user requests a web page from a web archive, the user will typically either get an HTTP 200 if the page is available, or an HTTP 404 if the web page has not been archived. This is because web archives are typically accessed by URI lookup, and the response is binary: the archive either has the page or it does not, and the user will not know of other archived web pages that exist and are potentially similar to the requested web page. In this paper, we propose augmenting these binary responses with a model for selecting and ranking recommended web pages in a Web archive. This is to enhance both HTTP 404 responses and HTTP 200 responses by surfacing web pages in the archive that the user may not know existed. First, we check if the URI is already classified in DMOZ or Wikipedia. If the requested URI is not found, we use ML to classify the URI using DMOZ as our ontology and collect candidate URIs to recommended to the user. Next, we filter the candidates based on if they are present in the archive. Finally, we rank candidates based on several features, such as archival quality, web page popularity, temporal similarity, and URI similarity. We calculated the F1 score for different methods of classifying the requested web page at the first level. We found that using all-grams from the URI after removing numerals and the TLD produced the best result with F1=0.59. For second-level classification, the micro-average F1=0.30. We found that 44.89% of the correctly classified URIs contained at least one word that exists in a dictionary and 50.07% of the correctly classified URIs contained long strings in the domain. In comparison with the URIs from our Wayback access logs, only 5.39% of those URIs contained only words from a dictionary, and 26.74% contained at least one word from a dictionary. These percentages are low and may affect the ability for the requested URI to be correctly classified.
Tasks
Published	2019-08-07
URL	https://arxiv.org/abs/1908.02819v1
PDF	https://arxiv.org/pdf/1908.02819v1.pdf
PWC	https://paperswithcode.com/paper/making-recommendations-from-web-archives-for
Repo
Framework

Deep Reinforcement Learning for Chatbots Using Clustered Actions and Human-Likeness Rewards


Title	Deep Reinforcement Learning for Chatbots Using Clustered Actions and Human-Likeness Rewards
Authors	Heriberto Cuayáhuitl, Donghyeon Lee, Seonghan Ryu, Sungja Choi, Inchul Hwang, Jihie Kim
Abstract	Training chatbots using the reinforcement learning paradigm is challenging due to high-dimensional states, infinite action spaces and the difficulty in specifying the reward function. We address such problems using clustered actions instead of infinite actions, and a simple but promising reward function based on human-likeness scores derived from human-human dialogue data. We train Deep Reinforcement Learning (DRL) agents using chitchat data in raw text—without any manual annotations. Experimental results using different splits of training data report the following. First, that our agents learn reasonable policies in the environments they get familiarised with, but their performance drops substantially when they are exposed to a test set of unseen dialogues. Second, that the choice of sentence embedding size between 100 and 300 dimensions is not significantly different on test data. Third, that our proposed human-likeness rewards are reasonable for training chatbots as long as they use lengthy dialogue histories of >=10 sentences.
Tasks	Sentence Embedding
Published	2019-08-27
URL	https://arxiv.org/abs/1908.10331v1
PDF	https://arxiv.org/pdf/1908.10331v1.pdf
PWC	https://paperswithcode.com/paper/deep-reinforcement-learning-for-chatbots
Repo
Framework

Generalizable Resource Allocation in Stream Processing via Deep Reinforcement Learning


Title	Generalizable Resource Allocation in Stream Processing via Deep Reinforcement Learning
Authors	Xiang Ni, Jing Li, Mo Yu, Wang Zhou, Kun-Lung Wu
Abstract	This paper considers the problem of resource allocation in stream processing, where continuous data flows must be processed in real time in a large distributed system. To maximize system throughput, the resource allocation strategy that partitions the computation tasks of a stream processing graph onto computing devices must simultaneously balance workload distribution and minimize communication. Since this problem of graph partitioning is known to be NP-complete yet crucial to practical streaming systems, many heuristic-based algorithms have been developed to find reasonably good solutions. In this paper, we present a graph-aware encoder-decoder framework to learn a generalizable resource allocation strategy that can properly distribute computation tasks of stream processing graphs unobserved from training data. We, for the first time, propose to leverage graph embedding to learn the structural information of the stream processing graphs. Jointly trained with the graph-aware decoder using deep reinforcement learning, our approach can effectively find optimized solutions for unseen graphs. Our experiments show that the proposed model outperforms both METIS, a state-of-the-art graph partitioning algorithm, and an LSTM-based encoder-decoder model, in about 70% of the test cases.
Tasks	Graph Embedding, graph partitioning
Published	2019-11-19
URL	https://arxiv.org/abs/1911.08517v1
PDF	https://arxiv.org/pdf/1911.08517v1.pdf
PWC	https://paperswithcode.com/paper/generalizable-resource-allocation-in-stream
Repo
Framework

Visual Perception and Modelling in Unstructured Orchard for Apple Harvesting Robots


Title	Visual Perception and Modelling in Unstructured Orchard for Apple Harvesting Robots
Authors	Hanwen Kang, Chao Chen
Abstract	Vision perception and modelling are the essential tasks of robotic harvesting in the unstructured orchard. This paper develops a framework of visual perception and modelling for robotic harvesting of fruits in the orchard environments. The developed framework includes visual perception, scenarios mapping, and fruit modelling. The Visual perception module utilises a deep-learning model to perform multi-purpose visual perception task within the working scenarios; The scenarios mapping module applies OctoMap to represent the multiple classes of objects or elements within the environment; The fruit modelling module estimates the geometry property of objects and estimates the proper access pose of each fruit. The developed framework is implemented and evaluated in the apple orchards. The experiment results show that visual perception and modelling algorithm can accurately detect and localise the fruits, and modelling working scenarios in real orchard environments. The $F_{1}$ score and mean intersection of union of visual perception module on fruit detection and segmentation are 0.833 and 0.852, respectively. The accuracy of the fruit modelling in terms of centre localisation and pose estimation are 0.955 and 0.923, respectively. Overall, an accurate visual perception and modelling algorithm are presented in this paper.
Tasks	Pose Estimation
Published	2019-12-29
URL	https://arxiv.org/abs/1912.12555v1
PDF	https://arxiv.org/pdf/1912.12555v1.pdf
PWC	https://paperswithcode.com/paper/visual-perception-and-modelling-in
Repo
Framework

An Introduction to Advanced Machine Learning : Meta Learning Algorithms, Applications and Promises


Title	An Introduction to Advanced Machine Learning : Meta Learning Algorithms, Applications and Promises
Authors	Farid Ghareh Mohammadi, M. Hadi Amini, Hamid R. Arabnia
Abstract	In [1, 2], we have explored the theoretical aspects of feature extraction optimization processes for solving largescale problems and overcoming machine learning limitations. Majority of optimization algorithms that have been introduced in [1, 2] guarantee the optimal performance of supervised learning, given offline and discrete data, to deal with curse of dimensionality (CoD) problem. These algorithms, however, are not tailored for solving emerging learning problems. One of the important issues caused by online data is lack of sufficient samples per class. Further, traditional machine learning algorithms cannot achieve accurate training based on limited distributed data, as data has proliferated and dispersed significantly. Machine learning employs a strict model or embedded engine to train and predict which still fails to learn unseen classes and sufficiently use online data. In this chapter, we introduce these challenges elaborately. We further investigate Meta-Learning (MTL) algorithm, and their application and promises to solve the emerging problems by answering how autonomous agents can learn to learn?.
Tasks	Meta-Learning
Published	2019-08-26
URL	https://arxiv.org/abs/1908.09788v1
PDF	https://arxiv.org/pdf/1908.09788v1.pdf
PWC	https://paperswithcode.com/paper/an-introduction-to-advanced-machine-learning
Repo
Framework

Generic Encodings of Constructor Rewriting Systems


Title	Generic Encodings of Constructor Rewriting Systems
Authors	Horatiu Cirstea, Pierre-Etienne Moreau
Abstract	Rewriting is a formalism widely used in computer science and mathematical logic. The classical formalism has been extended, in the context of functional languages, with an order over the rules and, in the context of rewrite based languages, with the negation over patterns. We propose in this paper a concise and clear algorithm computing the difference over patterns which can be used to define generic encodings of constructor term rewriting systems with negation and order into classical term rewriting systems. As a direct consequence, established methods used for term rewriting systems can be applied to analyze properties of the extended systems. The approach can also be seen as a generic compiler which targets any language providing basic pattern matching primitives. The formalism provides also a new method for deciding if a set of patterns subsumes a given pattern and thus, for checking the presence of useless patterns or the completeness of a set of patterns.
Tasks
Published	2019-05-14
URL	https://arxiv.org/abs/1905.06233v2
PDF	https://arxiv.org/pdf/1905.06233v2.pdf
PWC	https://paperswithcode.com/paper/generic-encodings-of-constructor-rewriting
Repo
Framework

Graph-based Neural Sentence Ordering


Title	Graph-based Neural Sentence Ordering
Authors	Yongjing Yin, Linfeng Song, Jinsong Su, Jiali Zeng, Chulun Zhou, Jiebo Luo
Abstract	Sentence ordering is to restore the original paragraph from a set of sentences. It involves capturing global dependencies among sentences regardless of their input order. In this paper, we propose a novel and flexible graph-based neural sentence ordering model, which adopts graph recurrent network \cite{Zhang:acl18} to accurately learn semantic representations of the sentences. Instead of assuming connections between all pairs of input sentences, we use entities that are shared among multiple sentences to make more expressive graph representations with less noise. Experimental results show that our proposed model outperforms the existing state-of-the-art systems on several benchmark datasets, demonstrating the effectiveness of our model. We also conduct a thorough analysis on how entities help the performance.
Tasks	Sentence Ordering
Published	2019-12-16
URL	https://arxiv.org/abs/1912.07225v1
PDF	https://arxiv.org/pdf/1912.07225v1.pdf
PWC	https://paperswithcode.com/paper/graph-based-neural-sentence-ordering
Repo
Framework

Active Learning in Video Tracking


Title	Active Learning in Video Tracking
Authors	Sima Behpour
Abstract	Active learning methods, like uncertainty sampling, combined with probabilistic prediction techniques have achieved success in various problems like image classification and text classification. For more complex multivariate prediction tasks, the relationships between labels play an important role in designing structured classifiers with better performance. However, computational time complexity limits prevalent probabilistic methods from effectively supporting active learning. Specifically, while non-probabilistic methods based on structured support vector machines can be tractably applied to predicting bipartite matchings, conditional random fields are intractable for these structures. We propose an adversarial approach for active learning with structured prediction domains that is tractable for matching. We evaluate this approach algorithmically in an important structured prediction problems: object tracking in videos. We demonstrate better accuracy and computational efficiency for our proposed method.
Tasks	Active Learning, Image Classification, Object Tracking, Structured Prediction, Text Classification
Published	2019-12-29
URL	https://arxiv.org/abs/1912.12557v3
PDF	https://arxiv.org/pdf/1912.12557v3.pdf
PWC	https://paperswithcode.com/paper/active-learning-in-video-tracking
Repo
Framework

Natural Language Processing via LDA Topic Model in Recommendation Systems


Title	Natural Language Processing via LDA Topic Model in Recommendation Systems
Authors	Hamed Jelodar, Yongli Wang, Mahdi Rabbani, SeyedValyAllah Ayobi
Abstract	Today, Internet is one of the widest available media worldwide. Recommendation systems are increasingly being used in various applications such as movie recommendation, mobile recommendation, article recommendation and etc. Collaborative Filtering (CF) and Content-Based (CB) are Well-known techniques for building recommendation systems. Topic modeling based on LDA, is a powerful technique for semantic mining and perform topic extraction. In the past few years, many articles have been published based on LDA technique for building recommendation systems. In this paper, we present taxonomy of recommendation systems and applications based on LDA. In addition, we utilize LDA and Gibbs sampling algorithms to evaluate ISWC and WWW conference publications in computer science. Our study suggest that the recommendation systems based on LDA could be effective in building smart recommendation system in online communities.
Tasks	Recommendation Systems
Published	2019-09-20
URL	https://arxiv.org/abs/1909.09551v1
PDF	https://arxiv.org/pdf/1909.09551v1.pdf
PWC	https://paperswithcode.com/paper/natural-language-processing-via-lda-topic
Repo
Framework

Compositional Structure Learning for Sequential Video Data


Title	Compositional Structure Learning for Sequential Video Data
Authors	Kyoung-Woon On, Eun-Sol Kim, Yu-Jung Heo, Byoung-Tak Zhang
Abstract	Conventional sequential learning methods such as Recurrent Neural Networks (RNNs) focus on interactions between consecutive inputs, i.e. first-order Markovian dependency. However, most of sequential data, as seen with videos, have complex temporal dependencies that imply variable-length semantic flows and their compositions, and those are hard to be captured by conventional methods. Here, we propose Temporal Dependency Networks (TDNs) for learning video data by discovering these complex structures of the videos. The TDNs represent video as a graph whose nodes and edges correspond to frames of the video and their dependencies respectively. Via a parameterized kernel with graph-cut and graph convolutions, the TDNs find compositional temporal dependencies of the data in multilevel graph forms. We evaluate the proposed method on the large-scale video dataset Youtube-8M. The experimental results show that our model efficiently learns the complex semantic structure of video data.
Tasks
Published	2019-07-03
URL	https://arxiv.org/abs/1907.01709v1
PDF	https://arxiv.org/pdf/1907.01709v1.pdf
PWC	https://paperswithcode.com/paper/compositional-structure-learning-for
Repo
Framework

Mitigating large adversarial perturbations on X-MAS (X minus Moving Averaged Samples)


Title	Mitigating large adversarial perturbations on X-MAS (X minus Moving Averaged Samples)
Authors	Woohyung Chun, Sung-Min Hong, Junho Huh, Inyup Kang
Abstract	We propose the scheme that mitigates the adversarial perturbation $\epsilon$ on the adversarial example $X_{adv}$ ($=$ $X$ $\pm$ $\epsilon$, $X$ is a benign sample) by subtracting the estimated perturbation $\hat{\epsilon}$ from $X$ $+$ $\epsilon$ and adding $\hat{\epsilon}$ to $X$ $-$ $\epsilon$. The estimated perturbation $\hat{\epsilon}$ comes from the difference between $X_{adv}$ and its moving-averaged outcome $W_{avg}X_{adv}$ where $W_{avg}$ is $N \times N$ moving average kernel that all the coefficients are one. Usually, the adjacent samples of an image are close to each other such that we can let $X$ $\approx$ $W_{avg}X$ (naming this relation after X-MAS[X minus Moving Averaged Samples]). By doing that, we can make the estimated perturbation $\hat{\epsilon}$ falls within the range of $\epsilon$. The scheme is also extended to do the multi-level mitigation by configuring the mitigated adversarial example $X_{adv}$ $\pm$ $\hat{\epsilon}$ as a new adversarial example to be mitigated. The multi-level mitigation gets $X_{adv}$ closer to $X$ with a smaller (i.e. mitigated) perturbation than original unmitigated perturbation by setting the moving averaged adversarial sample $W_{avg} * X_{adv}$ (which has the smaller perturbation than $X_{adv}$ if $X$ $\approx$ $W_{avg}*X$) as the boundary condition that the multi-level mitigation cannot cross over (i.e. decreasing $\epsilon$ cannot go below and increasing $\epsilon$ cannot go beyond). With the multi-level mitigation, we can get high prediction accuracies even in the adversarial example having a large perturbation (i.e. $\epsilon$ $>$ $16$). The proposed scheme is evaluated with adversarial examples crafted by the FGSM (Fast Gradient Sign Method) based attacks on ResNet-50 trained with ImageNet dataset.
Tasks
Published	2019-12-19
URL	https://arxiv.org/abs/1912.12170v4
PDF	https://arxiv.org/pdf/1912.12170v4.pdf
PWC	https://paperswithcode.com/paper/mitigating-large-adversarial-perturbations-on
Repo
Framework

Multiple instance learning with graph neural networks


Title	Multiple instance learning with graph neural networks
Authors	Ming Tu, Jing Huang, Xiaodong He, Bowen Zhou
Abstract	Multiple instance learning (MIL) aims to learn the mapping between a bag of instances and the bag-level label. In this paper, we propose a new end-to-end graph neural network (GNN) based algorithm for MIL: we treat each bag as a graph and use GNN to learn the bag embedding, in order to explore the useful structural information among instances in bags. The final graph representation is fed into a classifier for label prediction. Our algorithm is the first attempt to use GNN for MIL. We empirically show that the proposed algorithm achieves the state of the art performance on several popular MIL data sets without losing model interpretability.
Tasks	Multiple Instance Learning
Published	2019-06-12
URL	https://arxiv.org/abs/1906.04881v1
PDF	https://arxiv.org/pdf/1906.04881v1.pdf
PWC	https://paperswithcode.com/paper/multiple-instance-learning-with-graph-neural
Repo
Framework